Gameplay will only become more diversified if there's a reason to use different units. I don't know why you seem to think that tankspam is the way the game was meant to be played. Instead, the choice of units should be based upon the situation and as a direct response to the information that you've scouted.
Because in the future, there will be a T1 Riot Bot, with long range, high speed, large AoE/projectile count, but shitty health and poor raw DPS. It would be able to out-kite Tankblobs and fire from behind terrain cover, but would loose out to a flanking, pincer or backstabbing maneuver. Or an air strike. Counters are to RTS games as diarrhea is to Taco Bell. Entirely expected and incomplete without.
Obviously you have to scout but at the end of the day mass production is the way the game is meant to be played. Tanks are a direct damage land unit in a game of territory control. Most of the maps have most of the metal on land so of course tanks are required. Tanks can't win on their own but just like in supcom and TA they will always be present in large amounts. Opponent turtling? Better get some tanks to encricle him/her or crack the shell. Opponent building lots of air? Better build some interceptors and AA to support my roving parties of tanks which need to cut off his/her resources or destroy the base itself. Opponent spamming bots? Better build some bots to counter his raiding while rolling tanks towards his base forcing him to engage the tank blob. It's the same deal with air and navy. There will always be huge amounts of subs/frigates and interceptors to gain control of their respective theatres. Anyway "spam" in a game whose main mechanic is massive production is just meaningless. In most of the games in which people are accused of "tank spam" the opponent neglects the fact that the "spammer" often has more resources , more air factories and more bot factories than the opponent. They've just been outproduced.
And without some form of counterplay option the game becomes about who has the biggest economy and the best logistics management, rather than anything to do with strategic thought. It's telling that every one of your examples seems to have just one 'correct' reaction.
I just think that something needs to be pointed out here. The overall economic goal of all players is fairly simple, and should remain so - convert metal into military units at the highest rate possible. The resources required to do this are the metal itself, the build capacity to do the resources, and the energy to run the build capacity. This means that the player is always stalling on one of the three resources - metal, energy or build capacity. Balancing these to limit the stall, is the sign of a well balanced economy. At the moment, with the exception of a handful of games, metal is rarely the limiting resource. In the early game it tends to be energy, and in the late game, it tends to be build capacity. By extension, this means that players who are currently classed as good, play best by minimising their stall rate, and constantly increasing their build capacity as games continue. Hence why the maxim of "never stop expanding" remains true, and many players feel they are force to constantly build more factories, and spam out waves of units. We all want this game to be about large quantities of units, but this present economy is extremely rewarding to those who are able to maintain constant construction, without getting sidetracked. Hence why things feel a bit spammy. Metal should be the limiting resource more often, as it is the one most directly tied to territory. Thus there is a direct feedback of rewarding those who can capture and hold more territory, instead of rewarding those who can fill up their territory with P-gens and factories the fastest. This will change if the amount of metal income drops. The amount of metal needs to drop, not just by any amount, but by the amount required for it to become the limiting resource in most scenarios. Not all, as this would remove texture from the gameplay, but enough that it becomes a bit more scarce. I reckon doing this will shift a little more focus to military strategy, and away from the constant requirement to build more factories, without removing it completely. I expect this to happen in the fullness of time, sometime around the point where most of the unit diversity has been fleshed out. As always, we are forced to confront the fact that balance just isn't done yet. I know it's not a satisfactory mantra, but it's the one that's true.
Honestly, MadSci, I think it feels spammy at the moment because we don't have much unit variety. I'm reserving judgement until we have a much larger sample of the final unit pool. Also, I believe the intention is to significantly reduce metal availability once they've worked out how to get it to appear more 'intelligently' near player starting locations. Currently they just stuck a load down to make sure everyone has at least some access to it.
Unit variety would help. Actually needing metal would help. Currently you can build pretty large at full production spamming instant units with ease. Your economy stalling before you can get to build as fast as you do currently would help, and also make you have to "manage your money" better. On that subject, the devs need to hire an economy major, and not only have him scale things to where you only get a reasonable unit production speed with the metal available and speed-per-single-factory, but have him find the magic number that a t2 mex would have to produce in order to cripple the builder's economy and still not give infinite metal afterwards but give a big enough return to be stronger in the endgame due to it. At least that would also tie t2 mexes into dedicating playstyle to the early rush or endgame and thus weakening one's stance in the other. Generally though, making what you build matter only requires those two things, unit variety so some units you build for a single purpose and then you need some other units to defend against the purposes he does not fulfill, and limited economy so you only have so much metal to produce units with. The later point, a big help would not be to just limit the metal lower, but generally limit the whole effect of economy, where a marginally bad economy doesn't make you 1/2 the opponent's strength and a very good economy doesn't make you 3x stronger, but instead reguardless of economy everyone is strong enough to put up a fight and an economy gives you a smaller but still effective edge. Lowering it is a good start torwards it, because it brings the high closer to the low, while the egg helps because it starts everyone off with a boost so the weaker player starts higher than he would have by time the stronger player was at the same spot. Unit variety promises to be epic enough without comment. Naval functioning, variants of speed and such, different ranges.